Because retroviruses like HIV-1 evolve so rapidly, viral genetic sequences retrieved from just a few decades before present are effectively molecular 'fossils'. Such fossil sequences can be used to provide indisputable evidence of the existence of a given microbe at an early time point, and to validate historical inferences that could otherwise not be tested. They light up portions of the deep past that would otherwise be lost, and give us the means to test critically important hypotheses about the origins and evolutionary trajectories of medically important viruses. Although the battle to control HIV will always require the best possible understanding of its current genetic diversity, such understanding can only be achieved by viewing contemporary gene sequences in the context of historical genetic variation. With just a handful of notable exceptions the 'molecular archeology' perspective has to date not been exploited with respect to HIV-1 group M, the focus of this proposal. The overarching aim of this project is to make a comprehensive and integrated effort to understand the future of HIV-1 group M by recovering then analyzing information gleaned from its past. To do this we will 1. use archival specimens to uncover the history of the emergence of HIV in North America to determine whether Haiti was the "stepping stone" of HIV-1 group M subtype B before it spread around the developed world; and 2. investigate the evolution of HIV-1 in central Africa at an unprecedented early time point by optimizing DMA extraction protocols, identifying high-sensitivity PCR screening primers, exploring methods for pooling samples, and screening a small subset of our archived African tissue specimens. Crucially, we have assembled a massive collection of archival materials, many from sources that have not yet been widely considered in the field of HIV research, with which we will illuminate the remarkably unstudied early phase of HIV evolution (from around the time of the discovery of AIDS, back to 1955 and perhaps even earlier).